Usually, to search desired information or web sites, people can connect to a portal site and input search words in order to search relevant information or web sites.
Here, input query language need not to be words but may be characters or classification numbers.
On the other hand, in the case of the conventional search methods, query words input on a search window are parsed by a word, and then web pages on a web are indexed and displayed using a search method that uses operators of “AND,” “OR,” “NOT” etc., and similar query words are provided. In addition, correlation between words is calculated on the basis of a history or a collected document, and then several similar words are displayed using a visual method.
In addition, words connected with query words and a word degree of association are provided and a number of referenced relevant words are provided. However, if a relevant word is clicked, the clicked relevant word is searched again to thus progress a search work again. After such relevant words are searched, texts or images which are linked on a web page provided by an ordinary search engine site in the form of a hypertext markup language (HTML) type are selected to thus gain access to a relevant web page.
However, in the above-described conventional art, one web site that a user thinks to be the most similar web site is clicked among the.HTML list after search, to thus gain access to the web site. Accordingly, a majority of clicking times and one time of search are needed. As a result, there is a problem in the conventional art that much time is required because several times of clicks or enter key input processes are needed even if a search window has been maintained.
There is another method of connecting directly with a home page in which a mother language is input on a web browser address window and then collected URL addresses are returned on the address window. However, since a URL address is displayed instead of the input query words, there is a problem that query words cannot be used. Further, since there is a limitation that one word corresponds to one web site, there is a need to improve the conventional search method which is insufficient to gain access to various sub-pages.
Further, in the case of proper nouns, names of goods, registered trademarks, or enterprises that use similar company names, prior occupation of domains may cause a problem of raising an adverse effect that prevents competition in good faith.
Still further, when query words that are input on a search window are foreign language, a search work is performed using special keyboards or words that are found by connection of dictionary web sites on the Internet. However, a lot of times and efforts are needed. In particular, since several pages are opened at the same time to perform a search work, there is a problem that many loads are applied to a computer.